rusefi.com

Hello everyone.

First post here.

I Just wanted to introduce myself and get familiar with the forum here.

I've built a turbo setup on my otherwise stock nb miata, and ended up with RusEFI for a low cost controller solution

I ordered the Frankenso pre-build PnP controller, and will start playing around with it soon.

My goals for this project are to make reliable power on a mostly stock drivetrain(nothing crazy, just reliability).

I have a little background in fuel and spark control, but will likely be learning a lot more as I go down this path.

More to come...

Fun fact: I will be in Ann Arbor for the week a week from now

Welcome to the forum! Your next order of business is to spend some time with
https://rusefi.com/wiki/index.php?title=Vehicle:Mazda_Miata_1999
and https://rusefi.com/wiki/index.php?title=Manual:Hardware_Frankenso_board since you would need to cross-wire the board from NA6 to NB1 pinout. There are probably some threads on the forum with partial relevant hints.

Also there is https://github.com/rusefi/rusefi/blob/master/firmware/config/engines/mazda_miata_nb.cpp which is either helpful or distractful.

Welcome to rusefi!

Looks like you're running a weld-el manifold with a chinacharger?
Can't see your wastegate. What are youdoing for boost control?
What injectors are you running? NB Miatas have fixed fuel pressure, so the more boost you're running the faster you run out of fuel on a stock injector.

I have a (currently disassembled) 99 NB1 with an ancient FMII kit. I can upload you spark and fuel maps as starting points. However, I was running the flattop intake, so your stock intake will need less fuel at higher rpms.

Thanks! Ann Arbor is probaly on 30 minutes from me! Welcome to town
I wasn't aware that I will have to make some adjustements to make this work for NB, but no worries, I'm fine with soldering if required:)

Yes, the manifold is Schedule 40 mild steel weld-els.

The turbo is a "frankenturbo" with a turbonetics compressor housing, Aireasearch turbine housing, and unknown center housing cartridge(probably China)
I dont even know what size it is, but it seems ok, because it spools at just of 3k RPM(just the right amount of lag )

I'm running an external wastegate off the bottom of the manifold, but still need to merge it into my downpipe(finishing that tonight)
I will be using a manual boost controller run off the intake, and hopefully an electronic boost control eventually if RusEFI has the capability.

And I will be happy to take any maps from similar configurations to get me started!

Also, I'm running 460CC injectors from an FC RX7.

scramblr wrote: ↑

Fri Sep 20, 2019 8:04 pm

I wasn't aware that I will have to make some adjustements to make this work for NB, but no worries, I'm fine with soldering if required:)

https://rusefi.com/wiki/index.php?title=Manual:Hardware:Mazda_PnP

here are links to other NB1 installations
https://github.com/rusefi/rusefi/wiki/List_of_engines_running_rusEfi

I made some good progress over the weekend, an finished the hardware side of the of the turbo build.
I merged my wastegate, welded a flange for my BOV, and installed the 460CC injectors and MAP sensor.

But most importantly, the Frankenso ECU showed up in the mailbox!

I began following the prcocess of downloading/installing all the necessary programs and files, and after that quickly found out that I have a lot to learn!

There seems to be a wealth of information out there between this forum and the Wiki page, however the trouble is mining through it all and not knowing what I'm even looking for for.

I was able to connect the STM32 to the ST Link and program it with the latest rusefi.bin file.
Question: All the instructions I've found say to install the latest rusefi.hex file. There wasn't a hex file in the bundle I downloaded, so I'm assuming the binary file is ok, correct?

However, I wasn't able to connect to the Console or Tuner studio. After further reading, I found that I was using the wrong USB port but I have not verifed this yet.

Another question: The board shipped with 2 resistors. I found a thread about a possible issue with the IAC solenoid, but how will I know if I need to use these, and where would I install them if I do?

Yes, documentation is... not perfect.

Yes, .bin
Can you please do me a favor and point all places where "hex" is mentioned and I will edit?

These are diodes. You would 100% need one of these for your IAC fly-back, you can probably worry about this much much later? I will provide some documentation for that once we are closer

We put the documentation into the "ether" and it blew away in to the cosmos. Turns out the cloud and ether are two different things, who knew.
https://www.merriam-webster.com/dictionary/the%20ether

It would be great is better documentation existed. You tube video series like what Sydney Comic Hound have been doing are great. We need more of that.

Nice turbo.

I'm having a strange issue connecting to dev console.

I was able to connect fine via the micro usb port until I shorted PB1 to GND to allow connection to TS.

Ever since this change, I cannot seem to get dev console to connect.

I even erased the chip and reprogrammed in an attempt to reset whatever changed when I shorted PB1 to GND.

I am using windows 10 and found this thread that seemed to be relevant, but cannot determine what exactly I need to do to fix my issue.

https://rusefi.com/forum/viewtopic.php?f=5&t=263&p=17582&hilit=PB1#p17569

scramblr wrote: ↑

Tue Sep 24, 2019 12:27 am

I was able to connect fine via the micro usb port until I shorted PB1 to GND to allow connection to TS.

Where do we still mention PB1 to GND? this is long gone. Just use any port for any tool.

edited https://rusefi.com/forum/viewtopic.php?f=5&t=263&p=17569#p17569

This page is where I learned the trick, and it definitely worked because prior to shorting those pins I wasn't able to connect to TS.
https://rusefi.com/wiki/index.php?title=Manual:Software:TunerStudio_connectivity

scramblr wrote: ↑

Tue Sep 24, 2019 12:34 am

This page is where I learned the trick, and it definitely worked because prior to shorting those pins I wasn't able to connect to TS.
https://rusefi.com/wiki/index.php?title=Manual:Software:TunerStudio_connectivity

I've removed mentioning of PB1

microUSB port is easiest to get running since it does not care for baud rate. For mini USB port baud rate becomes important, I believe it could be 115000 by default now.

Update: I rebooted my PC and now can connect to both TS and Console via the micro USB port. No shorting PB1.

Should have tried this before posting for help but sometimes the obvious solution is the hardest to come by...

I actually think my issue is that TS and Console cannot be running at the same time.

If I'm connected to one of them, I'm not able to connect to the other.

I tested this by opening and connecting to TS and then opening Console and finding that I couldn't connect.

I closed both, and opened console and was able to connect.

I then opened TS and with console still connected, and couldn't connect to TS until I closed console.

I can't see why this should be an issue for me, but just wanted to point out my findings.

Sorry- I forgot about the promised tune files!
Here is a fuel tune and and aggressive (93 octane ONLY!) timing tune for turbo NB1. You should definitely drop the timing considerably. Maybe divide by 1.1 across the board to be safe. I had this running on a 99 head with a 94 bottom which drops the compression by about 0.5 points. And for some reason this engine resolutely refused to knock. So I just kept increasing the timing.

On the fuel tune I would recommend re-scaling the rpms before you auto-tune. The stock intake with the flaps that you operate (you do- right?) around 5000 rpm does funky things with the VE. When I had the stock intake I think I had rpm at like 4500, 4800, 5000, 5200 and 5500, just to get the needed resolution in that area.

I was running flowforce 650 injectors. Injector size doesn't really factor into the fuel table- it's supposed to be taken care of before you even get to the VE. I think your 400s are likely pretty lazy. So dead-time would be around 1.2ms - 1.4ms or so at 14V (just guessing here).

Get that baby running and idling and report back!

@stefanst, Thanks!

As far as repinning from NA to NB layout, I can't imagine that I'm the first to do it.
Has anybody compiled a reference guide for what pin on the current Frenkenso connector needs to be remapped for the NB connector?
If not, I will definitly compile this as I go through the process.
I have most of the info I need, but just want to see if anyone has anything that might save me a little time.

I did find this picture of an NB setup, which should help a little.
https://rusefi.com/forum/viewtopic.php?f=3&t=1282&hilit=99+miata&start=60#p28365

Another question:

In the picture I attached here, there are little components that connect the pin side of the connector to the board.
Are these simply jumpers, or are they a diode or something else?

The reason I ask is because when I go to repin my board for my NB miata, I want to know if I should remove this small jumper component from the pins that change.

these are 0R resistors, i.e. jumpers. just remove the ones which should not be jumping from board to connector

russian wrote: ↑

Tue Sep 24, 2019 5:54 pm

these are 0R resistors, i.e. jumpers. just remove the ones which should not be jumping from board to connector

Ok, good to know!

Question on adding MAP sensor and Wideband:
The current guidance I've seen on the wiki page and forum is that pins should be added to the stock harness at 3R and 3J https://rusefi.com/wiki/index.php?title=Manual:Hardware:Mazda_PnP
However, could I get the same result by wiring my MAP and Wideband to the pins currently used for the MAF and the narrowband sensors, which wont be used, and remap the inputs for MAP and wideband to the former inputs of MAF and narrowband in TS?

Very is _a lot_ of flexibility here. The only difference to note is that some analog inputs have 500K pull-down resistors and some have 10K pull-up resistors - you can see this on the back of the board. For analog sensor inputs you want to use channels with 500K or move 500K resistors.

Awesome! This makes me feel empowered!

Question on wiring the board for Cam and Crank sensor:

On the NB connectors, the signals for Cam and Crank come in at pins 2H and 2J.
On the NA connectors, the signals come it at 3E and 3J.
Instructions for wiring NA say to jumper 3E to 2B, and 3J to 2D. (https://rusefi.com/wiki/index.php?title=Vehicle:Mazda_Miata_1990)

So my question is, when I reroute the NB pins(2H and 2J), do I need to connect them to 3E and 3J, or can I just go straight ot 2B and 2D?

scramblr wrote: ↑

Wed Sep 25, 2019 11:12 pm

just go straight ot 2B and 2D?

because this is where the input is actually being decoded

Cool. So how do I handle the inputs from the NB harness that go pins 2A-2D(ex: IAT comes in at pin 2B)?
On the board, these are dedicated to cam and crank, but there aren't any jumpers to remove to reroute the other signals.
My only thought is that I need to physically cut the connector pins where they connect to the board and reroute them that way.
For example, cut pin 2B so that it can be routed to 3P which is the IAT signal location on the board.

Am I off base here, or is this the route I need to go?

Yes, 2A-B-C-D you would have to cut somewhere closer to the PCB. Bend the hanging part of the connector, leave some metal - probably not all 3cm and solder a wire to it. Wrap with heat shrink or tape to insulate.

Also, when I ordered the board you asked If my setup will measure crank position with a hall or VR sensor.
Does this mean the board is configured for my setup(Hall) specifically?

Yes, board is set for Hall.

Hall is easy to test on a bench - you simply GND input pins and trigger even counters in software would go up. Less easy with VR.

So now I've got the Cam and crank inputs to pins 2B and 2D squared away, but not sure what to do about pins 2A and 2C on the board.
Based on my analysis of the schematics it looks like these should both just get 12V from pin 1B.
Is this correct?